Spatial audio warp compensator

Abstract

Methods and devices for correcting warping in spatial audio may include identifying a geometric transform that defines a geometric warping between a first spatial geometric model that represents how sound is produced in a first volumetric space and a second spatial geometric model that represents how sound is produced in a second volumetric space different from the first volumetric space. The methods and devices may include determining an inverse of the geometric transform that compensates for the geometric transform. The methods and devices may include applying the inverse of the geometric transform to a first location in the first spatial geometric model by mapping the first location to a second location in the second spatial geometric model to correct for the geometric warping.

What is claimed is: 
     
       1. A computer device, comprising:
 a memory to store data and instructions; 
 a processor in communication with the memory; 
 an operating system in communication with the memory and the processor, wherein the operating system is operable to:
 identify a geometric transform that defines a geometric warping between a first spatial geometric model that represents how sound is produced in a first volumetric space and a second spatial geometric model that represents how sound is produced in a second volumetric space different from the first volumetric space, wherein the first spatial geometric model correlates to a normalized room geometry and the second spatial geometric model correlates to a physical room geometry; 
 determine an inverse of the geometric transform that compensates for the geometric transform; 
 compute a first mesh of the first spatial geometric model; 
 compute a second mesh of the second spatial geometric model; 
 determine a first face in the first mesh, wherein a volumetric space defined by the relationship of the first face to an origin contains a first location; 
 identify a second face in the second mesh corresponding to the first face in the first mesh; and 
 apply the inverse of the geometric transform to the first location in the first spatial geometric model by directly mapping the first location in the first face to a second location in the second face in the second spatial geometric model to correct for the geometric warping while maintaining a relative position of the first location in the first face and the second face to generate a dynamic audio object. 
 
 
     
     
       2. The computer device of  claim 1 , wherein the first spatial geometric model defines a first layout of static audio objects within the first volumetric space of a location relative to a fixed position, and
 wherein the second spatial geometric model defines a second layout of static audio objects within the second volumetric space relative to the fixed position. 
 
     
     
       3. The computer device of  claim 2 , wherein coordinates of the static objects in the first spatial geometric model are within a first range of positions, and
 wherein the coordinates of the static objects in the second spatial geometric model are within a second range of positions. 
 
     
     
       4. The computer device of  claim 2 , wherein the location is a physical room or a virtual room. 
     
     
       5. The computer device of  claim 2 , wherein the second spatial geometric model is based on requirements from an encoder. 
     
     
       6. The computer device of  claim 1 , wherein the operating system is further operable to:
 receive room geometry input of the second spatial geometric model from a user, wherein the room geometry input includes one or more of a location size, a number of static objects to place in the location, a number of dynamic objects to place in the room, locations of the static objects, and locations of the dynamic objects. 
 
     
     
       7. The computer device of  claim 1 , wherein the first spatial geometric model is based on one or more of an expected room geometry, a predicted room geometry, or a described room geometry of a rendering of encoded spatial audio. 
     
     
       8. The computer device of  claim 1 , wherein the operating system is further operable to determine the inverse of the geometric transform by placing a plurality of points into a prediction of the second spatial geometric model and transforming the plurality of points into the first spatial geometric model to calculate the inverse geometric transform. 
     
     
       9. A method for correcting warping in spatial audio, comprising:
 identifying, at an operating system executing on a computer device, a geometric transform that defines a geometric warping between a first spatial geometric model that represents how sound is produced in a first volumetric space and a second spatial geometric model that represents how sound is produced in a second volumetric space different from the first volumetric space, wherein the first spatial geometric model correlates to a normalized room geometry and the second spatial geometric model correlates to a physical room geometry; 
 determining, at the operating system, an inverse of the geometric transform that compensates for the geometric transform; 
 computing a first mesh of the first spatial geometric model; 
 computing a second mesh of the second spatial geometric model; 
 determining a first face in the first mesh, wherein a volumetric space defined by the relationship of the first face to an origin contains a first location; 
 identifying a second face in the second mesh corresponding to the first face in the first mesh; and 
 applying the inverse of the geometric transform to the first location in the first spatial geometric model by directly mapping the first location in the first face to a second location in the second face in the second spatial geometric model to correct for the geometric warping while maintaining a relative position of the first location in the first face and the second face to generate a dynamic audio object. 
 
     
     
       10. The method of  claim 9 , wherein the first spatial geometric model defines a first layout of static audio objects within the first volumetric space of a location relative to a fixed position, and
 wherein the second spatial geometric model defines a second layout of static audio objects within the second volumetric space relative to the fixed position. 
 
     
     
       11. The method of  claim 10 , wherein coordinates of the static objects in the first spatial geometric model are within a first range of positions, and
 wherein the coordinates of the static objects in the second spatial geometric model are within a second range of positions. 
 
     
     
       12. The method of  claim 10 , wherein the location is a physical room or a virtual room. 
     
     
       13. The method of  claim 10 , wherein the second spatial geometric model is based on requirements from an encoder. 
     
     
       14. The method of  claim 9 , further comprising:
 receiving room geometry input of the second spatial geometric model from a user, wherein the room geometry input includes one or more of a location size, a number of static objects to place in the location, a number of dynamic objects to place in the room, locations of the static objects, and locations of the dynamic objects. 
 
     
     
       15. The method of  claim 9 , wherein the first spatial geometric model is based on one or more of an expected room geometry, a predicted room geometry, or a described room geometry of a rendering of encoded spatial audio. 
     
     
       16. The method of  claim 9 , wherein determining the inverse of the geometric transform further comprises:
 placing a plurality of points into a prediction of the second spatial geometric model; and 
 transforming the plurality of points into the first spatial geometric model to calculate the inverse geometric transform. 
 
     
     
       17. A non-transitory computer-readable medium storing instructions executable by a computer device, comprising:
 at least one instruction for causing the computer device to identify a geometric transform that defines a geometric warping between a first spatial geometric model that represents how sound is produced in a first volumetric space and a second spatial geometric model that represents how sound is produced in a second volumetric space different from the first volumetric space, wherein the first spatial geometric model correlates to a normalized room geometry and the second spatial geometric model correlates to a physical room geometry; 
 at least one instruction for causing the computer device to determine an inverse of the geometric transform that compensates for the geometric transform; 
 at least one instruction for causing the computer device to compute a first mesh of the first spatial geometric model; 
 at least one instruction for causing the computer device to compute a second mesh of the second spatial geometric model; 
 at least one instruction for causing the computer device to determine a first face in the first mesh, wherein a volumetric space defined by the relationship of the first face to an origin contains a first location; 
 at least one instruction for causing the computer device to identify a second face in the second mesh corresponding to the first face in the first mesh; and 
 at least one instruction for causing the computer device to apply the inverse of the geometric transform to the first location in the first spatial geometric model by directly mapping the first location in the first face to a second location in the second face in the second spatial geometric model to correct for the geometric warping while maintaining a relative position of the first location in the first face and the second face to generate a dynamic audio object. 
 
     
     
       18. The method of  claim 9 , further comprising:
 defining a first line in the first mesh from the origin to an intersection point in the first mesh; 
 computing a first face of the first mesh by using the first line and the intersection point; 
 identifying a second face in the second mesh that directly corresponds to the first face of the first mesh; and 
 mapping the first location in the first face to a new mapped intersection location in the second face. 
 
     
     
       19. The method of  claim 18 , wherein mapping the first location to the new mapped intersection location further includes applying a scale factor during the mapping. 
     
     
       20. The computer device of  claim 1 , wherein the normalized room geometry correlates to an authoring model and the physical room geometry correlates to a rendering model.